In recent years, a network, which has a communication form in which data are freely transmitted and received among arbitrary nodes that constitute the network, has come to be used actively.
Conventionally, a centralized processing type network that includes a server, which functions as a host and is positioned in the center, and terminals, each of which acts as a client individually accessing the host server, was mainly used. Therefore, in the case where the terminals needed to communicate with each other, the communication is performed via the host server.
Meanwhile, a so-called distributed processing type network is gradually appearing. In order to realize a mechanism that distributes stored information or distributes process of the information, the communication function must also be distributed. Namely, data communications must be freely performed between each node that constitutes the network.
As a typical form, there is a form of a communication network called P2P (Peer to Peer). P2P is a form of use of the network in which the information is directly exchanged among a large unspecified number of nodes. There are two kinds of P2P. They are P2P that technically requires mediation of a central server and P2P that carries data in a bucket brigade type.
Also in the case where a central server is required, the central server only provides a file search database and performs connection management of the nodes. The exchange of data itself is performed by the direct connection between the concerned nodes.
A technology to efficiently realize a distributed-processing network form mentioned above has been studied (refer to patent documents 1 and 2). In both patent documents 1 and 2, the P2P system is used. Further, in Patent documents 1 and 2, there is formed a system in which the data are distributed to and shared with arbitrary nodes and the transmission and reception of data is performed.
Thereby, the flexibility as a usage form of the network system has improved, and users have gained a large convenience. However, on the other hand, since processing capability has to be distributed among the nodes, there may be a case where the distribution of the processing capability can be seen as a burden from the viewpoint of efficiency.
For example, in the case where data are distributed, a user is able to share a larger amount of data as the network scale becomes larger. However, in the conventional way, the user cannot acquire data only by accessing a specific server. It is also necessary to collect the required data distributed from a large network and restore the data in a required form.
It will be important from now how to reduce the burden accompanying such distributed processing.    Patent document 1: Unexamined Japanese Patent Application Publication No. 2005-275937    Patent document 2: Unexamined Japanese Patent Application Publication No. 2004-252732